Intelligent subway monitoring system

Abstract

The invention relates to metro monitoring, in particular to an intelligent metro monitoring system and a vehicle-mounted terminal, vehicle data is collected through a sensor module, meanwhile, corresponding position information when the sensor module obtains the vehicle data is collected through a positioning module, and a passenger distribution coefficient calculation module is used for calculating a passenger distribution coefficient of a vehicle according to the vehicle data; the passenger distribution coefficient, the vehicle data and the corresponding position information are packaged and then sent to a ground base station; the control terminal receives the data transmitted by the ground base station, processes the received data through a data processing module, evaluates the passenger flow and the vehicle running condition according to the processed received data by using a control decision module, formulates a control scheduling strategy, and sends the control scheduling strategy to the vehicle-mounted terminal through the ground base station; according to the technical scheme provided by the invention, the defect that accurate and efficient control scheduling is difficult to carry out on the metro depot in the prior art can be effectively overcome.

Description

Technical Field

[0001] The present invention relates to subway monitoring, and in particular to an intelligent subway monitoring system. Background Art

[0002] With the rapid increase of urban population, traffic congestion in big cities has followed. The original basic transportation facilities can no longer meet the current travel needs. The main way to solve this problem is to build underground railway facilities. Due to its advantages in transporting large passenger volume, fast and punctual operation, green and low-carbon travel, etc., the subway can solve the problem of insufficient capacity of traditional public transportation and effectively alleviate traffic pressure. It has been vigorously promoted and constructed in many cities and is also an important solution to solve the problem of traffic congestion in large cities in China.

[0003] At present, the degree of automation of subway control and dispatching is low. Dispatchers at various positions in the base mainly rely on manual command and dispatch, and control and dispatch plans are manually compiled and communicated verbally. Various systems are set up in a scattered manner, and the linkage between systems is limited, making it difficult to accurately and efficiently control and dispatch subway depots. Summary of the invention

[0004] 1. Technical issues to be solved

[0005] In view of the above-mentioned shortcomings of the prior art, the present invention provides a smart subway monitoring system, which can effectively overcome the defect of the prior art that it is difficult to accurately and efficiently control and dispatch the subway vehicle depot.

[0006] (II) Technical solution

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0008] A smart subway monitoring system includes an on-board terminal and a control terminal;

[0009] The vehicle terminal collects vehicle data through the sensor module, and collects the corresponding position information when the sensor module obtains the vehicle data through the positioning module, calculates the passenger distribution coefficient of the vehicle according to the vehicle data using the passenger distribution coefficient calculation module, and encapsulates the passenger distribution coefficient, vehicle data and its corresponding position information and sends them to the ground base station;

[0010] The control terminal receives data transmitted by the ground base station, processes the received data through the data processing module, uses the control decision module to evaluate the passenger flow and vehicle operation status according to the processed received data, and formulates a control scheduling strategy, which is sent to the vehicle terminal through the ground base station;

[0011] Among them, the data processing module uses improved non-negative matrix factorization (NMF) to fill in the missing values ​​in the passenger distribution coefficient and vehicle data. The improved non-negative matrix factorization (NMF) improves the accuracy of the filled data by introducing a matrix that represents the missing value data structure, while maintaining the consistency of the filled data with the original data.

[0012] Preferably, the vehicle terminal collects vehicle data through the sensor module, and collects the corresponding position information when the sensor module obtains the vehicle data through the positioning module, including:

[0013] The vehicle terminal collects vehicle data through sensor modules installed inside each compartment. The vehicle data includes the heat source area and the number of heat source targets collected by infrared sensors in each compartment, and the operating status data of each compartment collected by other sensors;

[0014] At the same time, the on-board terminal collects the corresponding location information when the sensor module obtains vehicle data through the positioning module installed inside each car. The location information includes the car number, the nearest station information and the GPS location information.

[0015] Preferably, the vehicle-mounted terminal calculates the passenger distribution coefficient of the vehicle according to the vehicle data using the passenger distribution coefficient calculation module, including:

[0016] The vehicle terminal uses the passenger distribution coefficient calculation module to calculate the passenger distribution coefficient of the vehicle according to the heat source area and the number of heat source targets in each compartment using the following formula:

[0017]

[0018] Among them, PDC i is the passenger distribution coefficient of vehicle i, S ij , S' ij are the heat source area and the maximum passenger area of ​​compartment j in vehicle i, respectively. ij 、N' ij are the target number of heat sources and the maximum number of passengers in compartment j of vehicle i, respectively, j is the weight coefficient of carriage j in vehicle i, which is set according to the functional attributes of the building facilities around the vehicle arrival station, j∈[1,n], and n is the number of carriages in vehicle i.

[0019] Preferably, the vehicle-mounted terminal encapsulates the passenger distribution coefficient, vehicle data and their corresponding location information and sends them to the ground base station, including:

[0020] The vehicle terminal encapsulates the passenger distribution coefficient, vehicle data and its corresponding location information, and sends them to the ground base station through the wireless communication module;

[0021] After receiving the encapsulated data, the ground base station decodes and verifies it to ensure the accuracy and integrity of the data, and transmits the data to the control terminal.

[0022] Preferably, the control terminal receives data transmitted by the ground base station, and processes the received data through the data processing module, including:

[0023] The control terminal receives the data transmitted by the ground base station and pre-processes the received data through the data processing module;

[0024] The data processing module uses improved non-negative matrix factorization (NMF) to fill in the missing values ​​in the pre-processed passenger distribution coefficient and vehicle data. The improved non-negative matrix factorization (NMF) improves the accuracy of the filled data by introducing a matrix that represents the missing value data structure, while maintaining the consistency of the filled data with the original data.

[0025] The data processing module pre-processes the received data including removing outliers and smoothing.

[0026] Preferably, the data processing module uses improved non-negative matrix factorization (NMF) to fill in missing values ​​in the preprocessed passenger distribution coefficient and vehicle data, including:

[0027] S1, organize the preprocessed passenger distribution coefficient and vehicle data into a non-negative matrix V;

[0028] Among them, the missing values ​​in the non-negative matrix V are represented by 0, each row of the matrix represents a time point, and each column of the matrix represents a data category, which includes the heat source area of ​​each compartment, the number of heat source targets, the operating status data, and the passenger distribution coefficient of the vehicle;

[0029] S2, introduce a non-negative matrix Z that represents the missing value data structure, adjust the objective function of the NMF model according to the non-negative matrix Z, and determine the corresponding iterative optimization strategy;

[0030] Among them, in the non-negative matrix Z, 1 is used to represent missing values, and 0 is used to represent non-missing values;

[0031] S3, solving the NMF model based on the iterative optimization strategy. In each iteration, other matrices are fixed and one matrix is ​​updated until the iteration termination condition is met, and the decomposed non-negative matrix W, non-negative matrix H and non-negative matrix Z are obtained;

[0032] Among them, the non-negative matrix W is the basic image matrix, which contains the features or basis extracted from the non-negative matrix V, and each column of the matrix is ​​regarded as a basic feature or basis vector;

[0033] The non-negative matrix H is a coefficient matrix, which contains coefficients for linearly combining basis vectors in the non-negative matrix W. Each row of the matrix corresponds to a data point in the non-negative matrix V, and each column of the matrix corresponds to a basis vector in the non-negative matrix W.

[0034] S4, filling missing values ​​in the preprocessed passenger distribution coefficient and vehicle data according to the decomposed non-negative matrix W, non-negative matrix H and non-negative matrix Z, and calculating the filling data corresponding to the missing values;

[0035] S5. Smooth and verify the calculated filling data to ensure the accuracy and rationality of the filling data.

[0036] Preferably, a non-negative matrix Z representing the missing value data structure is introduced in S2, the objective function of the NMF model is adjusted according to the non-negative matrix Z, and a corresponding iterative optimization strategy is determined, including:

[0037] S21, determine the data structure of missing values, and introduce a non-negative matrix Z that represents the data structure of missing values;

[0038] S22. Adjust the objective function of the NMF model according to the non-negative matrix Z:

[0039]

[0040] Among them, V, W, H, and Z are non-negative matrices V, W, H, and Z, respectively. nmv , H nmv are the non-missing values ​​observed in the non-negative matrix V and the non-negative matrix H, respectively, λ is the regularization parameter, represents the Hadamard product, represents the square of the Frobenius norm;

[0041] S23. Use the alternating minimization strategy as an iterative optimization strategy for solving the NMF model.

[0042] Preferably, in S3, the NMF model is solved based on an iterative optimization strategy. In each iteration, other matrices are fixed and one matrix is ​​updated until the iteration termination condition is met, and the decomposed non-negative matrix W, non-negative matrix H and non-negative matrix Z are obtained, including:

[0043] S31, solving the NMF model based on the alternating minimization strategy;

[0044] S32, in each iteration process, the non-negative matrix W, the non-negative matrix H, the non-negative matrix Z, and the padding data are updated in the order of updating;

[0045] S33, judging whether the iteration termination condition is satisfied, if not, returning to S31, otherwise taking the current non-negative matrix W, non-negative matrix H and non-negative matrix Z as the decomposed non-negative matrix W, non-negative matrix H and non-negative matrix Z;

[0046] The iteration termination conditions include the objective function value converging to below a certain threshold, or the number of iterations reaching a preset maximum number.

[0047] Preferably, the control terminal uses the control decision module to evaluate the passenger flow and vehicle operation according to the processed received data, and formulates a control scheduling strategy, and sends the control scheduling strategy to the vehicle terminal through the ground base station, including:

[0048] The control terminal uses the control decision module to evaluate the passenger flow based on the processed passenger distribution coefficient and the heat source area and the number of heat source targets collected by the infrared sensor in the vehicle data in each compartment;

[0049] The control terminal uses the control decision module to input the operating status data of each compartment collected by other sensors in the processed vehicle data, as well as the corresponding position information when other sensors obtain the operating status data, into the pre-trained vehicle operation status judgment model to evaluate the vehicle operation status;

[0050] The control decision module conducts a comprehensive analysis of the passenger flow assessment results and the vehicle operation assessment results, and formulates a control scheduling strategy. The control terminal sends the control scheduling strategy to the on-board terminal through the ground base station.

[0051] Preferably, the control terminal uses the control decision module to input the operating status data of each compartment collected by other sensors in the processed vehicle data, and the corresponding position information when other sensors obtain the operating status data into the pre-trained vehicle operation status judgment model, before evaluating the vehicle operation status, including:

[0052] S1. Divide the historical data set into a training set, a validation set, and a test set according to a preset ratio;

[0053] S2. Setting the loss function and optimizer of the vehicle operation condition judgment model;

[0054] S3, inputting the training set into the vehicle operation condition judgment model for model training;

[0055] S4, the loss value is calculated based on the loss function, and the optimizer updates the model parameters according to the loss value and network gradient information;

[0056] S5. If the loss value is less than the preset threshold, the model training ends, and the current vehicle operation condition judgment model is the trained vehicle operation condition judgment model. Otherwise, return to S3 and continue to use the training set for model training.

[0057] S6. Input the validation set into the trained vehicle operation condition judgment model, evaluate the generalization ability of the model by observing the performance on the validation set, and tune the hyperparameters and structure of the model;

[0058] S7. Input the test set into the optimized vehicle operation condition judgment model to evaluate the model performance.

[0059] (III) Beneficial effects

[0060] Compared with the prior art, the intelligent subway monitoring system provided by the present invention has the following beneficial effects:

[0061] 1) The control terminal receives data transmitted by the ground base station. The received data is first preprocessed by the data processing module, and then the data processing module uses the improved non-negative matrix factorization NMF to fill the missing values ​​in the preprocessed passenger distribution coefficient and vehicle data. The improved non-negative matrix factorization NMF introduces a matrix that represents the missing value data structure to improve the accuracy of the filled data, while maintaining the consistency of the filled data with the original data, thereby achieving accurate filling of missing values ​​in the time-series-based vehicle data, and providing data support for the subsequent control decision module to formulate control scheduling strategies;

[0062] 2) The sensor module collects vehicle data, the positioning module collects the corresponding position information when the sensor module obtains the vehicle data, the passenger distribution coefficient calculation module calculates the passenger distribution coefficient of the vehicle based on the vehicle data, the on-board terminal encapsulates the passenger distribution coefficient, vehicle data and its corresponding position information and sends them to the ground base station, the control terminal receives the data transmitted by the ground base station, and after the data is processed by the data processing module, accurate and complete received data can be obtained. The control decision module can realize accurate evaluation of passenger flow based on the processed passenger distribution coefficient and the heat source area and the number of heat source targets collected by the infrared sensor in the vehicle data in each compartment;

[0063] 3) At the same time, the control decision module inputs the operating status data of each car collected by other sensors in the processed vehicle data, as well as the corresponding position information when other sensors obtain the operating status data, into the pre-trained vehicle operation status judgment model, which can realize accurate evaluation of the vehicle operation status. The control decision module conducts a comprehensive analysis of the passenger flow evaluation results and the vehicle operation status evaluation results, and formulates a control and scheduling strategy to ensure the accuracy and rationality of the control and scheduling strategy, so as to accurately and efficiently control and schedule the subway vehicle depot. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0065] Figure 1 It is a schematic diagram of the system of the present invention;

[0066] Figure 2 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0067] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0068] A smart subway monitoring system, such as Figure 1 and Figure 2 As shown, it includes a vehicle-mounted terminal and a control terminal;

[0069] The vehicle terminal collects vehicle data through the sensor module, and collects the corresponding position information when the sensor module obtains the vehicle data through the positioning module, calculates the passenger distribution coefficient of the vehicle according to the vehicle data using the passenger distribution coefficient calculation module, and encapsulates the passenger distribution coefficient, vehicle data and its corresponding position information and sends them to the ground base station;

[0070] The control terminal receives data transmitted by the ground base station, processes the received data through the data processing module, uses the control decision module to evaluate the passenger flow and vehicle operation status according to the processed received data, and formulates a control scheduling strategy, which is sent to the vehicle terminal through the ground base station;

[0071] Among them, the data processing module uses improved non-negative matrix factorization (NMF) to fill in the missing values ​​in the passenger distribution coefficient and vehicle data. The improved non-negative matrix factorization (NMF) improves the accuracy of the filled data by introducing a matrix that represents the missing value data structure, while maintaining the consistency of the filled data with the original data.

[0072] In the above technical solution, the control terminal receives data transmitted by the ground base station, first pre-processes the received data through the data processing module, and then the data processing module uses the improved non-negative matrix decomposition NMF to fill the missing values ​​in the pre-processed passenger distribution coefficient and vehicle data. The improved non-negative matrix decomposition NMF improves the accuracy of the filled-in data by introducing a matrix that represents the data structure of the missing values, while maintaining the consistency of the filled-in data with the original data, thereby achieving accurate filling of the missing values ​​in the time series-based vehicle data, and providing data support for the subsequent control decision module to formulate a control scheduling strategy.

[0073] In the technical solution of this application, for the vehicle side:

[0074] ① The vehicle terminal collects vehicle data through the sensor module, and collects the corresponding location information when the sensor module obtains vehicle data through the positioning module, including:

[0075] The vehicle terminal collects vehicle data through sensor modules installed inside each compartment. The vehicle data includes the heat source area and the number of heat source targets collected by infrared sensors in each compartment, and the operating status data of each compartment collected by other sensors;

[0076] At the same time, the on-board terminal collects the corresponding location information when the sensor module obtains vehicle data through the positioning module installed inside each car. The location information includes the car number, the nearest station information and the GPS location information.

[0077] ② The vehicle terminal uses the passenger distribution coefficient calculation module to calculate the passenger distribution coefficient of the vehicle according to the vehicle data, including:

[0078] The vehicle terminal uses the passenger distribution coefficient calculation module to calculate the passenger distribution coefficient of the vehicle according to the heat source area and the number of heat source targets in each compartment using the following formula:

[0079]

[0080] Among them, PDC i is the passenger distribution coefficient of vehicle i, S ij , S' ij are the heat source area and the maximum passenger area of ​​compartment j in vehicle i, respectively. ij 、N' ij are the target number of heat sources and the maximum number of passengers in compartment j of vehicle i, respectively, j is the weight coefficient of carriage j in vehicle i, which is set according to the functional attributes of the building facilities around the vehicle arrival station, j∈[1,n], and n is the number of carriages in vehicle i.

[0081] ③ The vehicle terminal encapsulates the passenger distribution coefficient, vehicle data and its corresponding location information and sends them to the ground base station, including:

[0082] The vehicle terminal encapsulates the passenger distribution coefficient, vehicle data and its corresponding location information, and sends them to the ground base station through the wireless communication module;

[0083] After receiving the encapsulated data, the ground base station decodes and verifies it to ensure the accuracy and integrity of the data, and transmits the data to the control terminal.

[0084] In the technical solution of this application, for the control center:

[0085] ① The control terminal receives the data transmitted by the ground base station and processes the received data through the data processing module, including:

[0086] The control terminal receives the data transmitted by the ground base station and pre-processes the received data through the data processing module;

[0087] The data processing module uses improved non-negative matrix factorization (NMF) to fill in the missing values ​​in the pre-processed passenger distribution coefficient and vehicle data. The improved non-negative matrix factorization (NMF) improves the accuracy of the filled data by introducing a matrix that represents the missing value data structure, while maintaining the consistency of the filled data with the original data.

[0088] The data processing module pre-processes the received data including removing outliers and smoothing.

[0089] Specifically, the data processing module uses improved non-negative matrix factorization (NMF) to fill in the missing values ​​in the preprocessed passenger distribution coefficient and vehicle data, including:

[0090] S1, organize the preprocessed passenger distribution coefficient and vehicle data into a non-negative matrix V;

[0091] Among them, the missing values ​​in the non-negative matrix V are represented by 0, each row of the matrix represents a time point, and each column of the matrix represents a data category, which includes the heat source area of ​​each compartment, the number of heat source targets, the operating status data, and the passenger distribution coefficient of the vehicle;

[0092] S2, introduce a non-negative matrix Z that represents the missing value data structure, adjust the objective function of the NMF model according to the non-negative matrix Z, and determine the corresponding iterative optimization strategy;

[0093] Among them, in the non-negative matrix Z, 1 is used to represent missing values, and 0 is used to represent non-missing values;

[0094] S3, solving the NMF model based on the iterative optimization strategy. In each iteration, other matrices are fixed and one matrix is ​​updated until the iteration termination condition is met, and the decomposed non-negative matrix W, non-negative matrix H and non-negative matrix Z are obtained;

[0095] Among them, the non-negative matrix W is the basic image matrix, which contains the features or basis extracted from the non-negative matrix V, and each column of the matrix is ​​regarded as a basic feature or basis vector;

[0096] The non-negative matrix H is a coefficient matrix, which contains coefficients for linearly combining basis vectors in the non-negative matrix W. Each row of the matrix corresponds to a data point in the non-negative matrix V, and each column of the matrix corresponds to a basis vector in the non-negative matrix W.

[0097] S4, filling missing values ​​in the preprocessed passenger distribution coefficient and vehicle data according to the decomposed non-negative matrix W, non-negative matrix H and non-negative matrix Z, and calculating the filling data corresponding to the missing values;

[0098] S5. Smooth and verify the calculated filling data to ensure the accuracy and rationality of the filling data.

[0099] Specifically, a non-negative matrix Z representing the missing value data structure is introduced in S2. The objective function of the NMF model is adjusted according to the non-negative matrix Z, and the corresponding iterative optimization strategy is determined, including:

[0100] S21, determine the data structure of missing values, and introduce a non-negative matrix Z that represents the data structure of missing values;

[0101] S22. Adjust the objective function of the NMF model according to the non-negative matrix Z:

[0102]

[0103] Among them, V, W, H, and Z are non-negative matrices V, W, H, and Z, respectively. nmv , H nmv are the non-missing values ​​observed in the non-negative matrix V and the non-negative matrix H, respectively, λ is the regularization parameter, represents the Hadamard product, represents the square of the Frobenius norm;

[0104] S23. Use the alternating minimization strategy as an iterative optimization strategy for solving the NMF model.

[0105] Specifically, S3 solves the NMF model based on an iterative optimization strategy. In each iteration, other matrices are fixed and one matrix is ​​updated until the iteration termination condition is met. The decomposed non-negative matrix W, non-negative matrix H and non-negative matrix Z are obtained, including:

[0106] S31, solving the NMF model based on the alternating minimization strategy;

[0107] S32, in each iteration process, the non-negative matrix W, the non-negative matrix H, the non-negative matrix Z, and the padding data are updated in the order of updating;

[0108] S33, judging whether the iteration termination condition is satisfied, if not, returning to S31, otherwise taking the current non-negative matrix W, non-negative matrix H and non-negative matrix Z as the decomposed non-negative matrix W, non-negative matrix H and non-negative matrix Z;

[0109] The iteration termination conditions include the objective function value converging to below a certain threshold, or the number of iterations reaching a preset maximum number.

[0110] ② The control terminal uses the control decision module to evaluate the passenger flow and vehicle operation according to the processed received data, and formulates a control scheduling strategy, which is sent to the vehicle terminal through the ground base station, including:

[0111] The control terminal uses the control decision module to evaluate the passenger flow based on the processed passenger distribution coefficient and the heat source area and the number of heat source targets collected by the infrared sensor in the vehicle data in each compartment;

[0112] The control terminal uses the control decision module to input the operating status data of each compartment collected by other sensors in the processed vehicle data, as well as the corresponding position information when other sensors obtain the operating status data, into the pre-trained vehicle operation status judgment model to evaluate the vehicle operation status;

[0113] The control decision module conducts a comprehensive analysis of the passenger flow assessment results and the vehicle operation assessment results, and formulates a control scheduling strategy. The control terminal sends the control scheduling strategy to the on-board terminal through the ground base station.

[0114] Specifically, the control terminal uses the control decision module to input the operating status data of each compartment collected by other sensors in the processed vehicle data, as well as the corresponding position information when other sensors obtain the operating status data, into the pre-trained vehicle operation status judgment model, before evaluating the vehicle operation status, including:

[0115] S1. Divide the historical data set into a training set, a validation set, and a test set according to a preset ratio;

[0116] S2. Setting the loss function and optimizer of the vehicle operation condition judgment model;

[0117] S3, inputting the training set into the vehicle operation condition judgment model for model training;

[0118] S4, the loss value is calculated based on the loss function, and the optimizer updates the model parameters according to the loss value and network gradient information;

[0119] S5. If the loss value is less than the preset threshold, the model training ends, and the current vehicle operation condition judgment model is the trained vehicle operation condition judgment model. Otherwise, return to S3 and continue to use the training set for model training.

[0120] S6. Input the validation set into the trained vehicle operation condition judgment model, evaluate the generalization ability of the model by observing the performance on the validation set, and tune the hyperparameters and structure of the model;

[0121] S7. Input the test set into the optimized vehicle operation condition judgment model to evaluate the model performance.

[0122] In the above technical solution, the sensor module collects vehicle data, the positioning module collects the corresponding position information when the sensor module obtains the vehicle data, the passenger distribution coefficient calculation module calculates the passenger distribution coefficient of the vehicle according to the vehicle data, the on-board terminal encapsulates the passenger distribution coefficient, the vehicle data and its corresponding position information and sends them to the ground base station, the control terminal receives the data transmitted by the ground base station, and after the data is processed by the data processing module, accurate and complete received data can be obtained, and the control decision module can realize accurate evaluation of passenger flow according to the processed passenger distribution coefficient and the heat source area and the number of heat source targets of each compartment collected by the infrared sensor in the vehicle data;

[0123] At the same time, the control decision module inputs the operating status data of each car collected by other sensors in the processed vehicle data, as well as the corresponding position information when other sensors obtain the operating status data, into the pre-trained vehicle operation status judgment model, which can realize accurate evaluation of the vehicle operation status. The control decision module conducts a comprehensive analysis of the passenger flow evaluation results and the vehicle operation status evaluation results, and formulates a control and scheduling strategy to ensure the accuracy and rationality of the control and scheduling strategy, so as to accurately and efficiently control and schedule the subway vehicle depot.

[0124] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A smart subway monitoring system, characterized by: Including vehicle-mounted terminal and control terminal; The vehicle terminal collects vehicle data through the sensor module, and collects the corresponding position information when the sensor module obtains the vehicle data through the positioning module, calculates the passenger distribution coefficient of the vehicle according to the vehicle data using the passenger distribution coefficient calculation module, and encapsulates the passenger distribution coefficient, vehicle data and its corresponding position information and sends them to the ground base station; The control terminal receives data transmitted by the ground base station, processes the received data through the data processing module, uses the control decision module to evaluate the passenger flow and vehicle operation status according to the processed received data, and formulates a control scheduling strategy, which is sent to the vehicle terminal through the ground base station; Among them, the data processing module uses improved non-negative matrix factorization (NMF) to fill in the missing values ​​in the passenger distribution coefficient and vehicle data. The improved non-negative matrix factorization (NMF) improves the accuracy of the filled data by introducing a matrix that represents the missing value data structure, while maintaining the consistency of the filled data with the original data.

2. The intelligent subway monitoring system according to claim 1 is characterized in that: The vehicle terminal collects vehicle data through the sensor module, and collects the corresponding position information when the sensor module obtains the vehicle data through the positioning module, including: The vehicle terminal collects vehicle data through sensor modules installed inside each compartment. The vehicle data includes the heat source area and the number of heat source targets collected by infrared sensors in each compartment, and the operating status data of each compartment collected by other sensors; At the same time, the on-board terminal collects the corresponding location information when the sensor module obtains vehicle data through the positioning module installed inside each car. The location information includes the car number, the nearest station information and the GPS location information.

3. The intelligent subway monitoring system according to claim 2 is characterized in that: The vehicle-mounted terminal calculates the passenger distribution coefficient of the vehicle according to the vehicle data using the passenger distribution coefficient calculation module, including: The vehicle terminal uses the passenger distribution coefficient calculation module to calculate the passenger distribution coefficient of the vehicle according to the heat source area and the number of heat source targets in each compartment using the following formula: Among them, PDC i is the passenger distribution coefficient of vehicle i, S ij , S' ij are the heat source area and the maximum passenger area of ​​compartment j in vehicle i, respectively. ij 、N' ij are the target number of heat sources and the maximum number of passengers in compartment j of vehicle i, respectively, j is the weight coefficient of carriage j in vehicle i, which is set according to the functional attributes of the building facilities around the vehicle arrival station, j∈[1,n], and n is the number of carriages in vehicle i.

4. The intelligent subway monitoring system according to claim 3 is characterized in that: The vehicle-mounted terminal encapsulates the passenger distribution coefficient, vehicle data and corresponding location information and sends them to the ground base station, including: The vehicle terminal encapsulates the passenger distribution coefficient, vehicle data and its corresponding location information, and sends them to the ground base station through the wireless communication module; After receiving the encapsulated data, the ground base station decodes and verifies it to ensure the accuracy and integrity of the data, and transmits the data to the control terminal.

5. The intelligent subway monitoring system according to claim 4 is characterized in that: The control terminal receives data transmitted by the ground base station and processes the received data through the data processing module, including: The control terminal receives the data transmitted by the ground base station and pre-processes the received data through the data processing module; The data processing module uses improved non-negative matrix factorization (NMF) to fill in the missing values ​​in the pre-processed passenger distribution coefficient and vehicle data. The improved non-negative matrix factorization (NMF) improves the accuracy of the filled data by introducing a matrix that represents the missing value data structure, while maintaining the consistency of the filled data with the original data. The data processing module pre-processes the received data including removing outliers and smoothing.

6. The intelligent subway monitoring system according to claim 5 is characterized in that: The data processing module uses improved non-negative matrix factorization (NMF) to fill in missing values ​​in the pre-processed passenger distribution coefficient and vehicle data, including: S1, organize the preprocessed passenger distribution coefficient and vehicle data into a non-negative matrix V; Among them, the missing values ​​in the non-negative matrix V are represented by 0, each row of the matrix represents a time point, and each column of the matrix represents a data category, which includes the heat source area of ​​each compartment, the number of heat source targets, the operating status data, and the passenger distribution coefficient of the vehicle; S2, introduce a non-negative matrix Z that represents the missing value data structure, adjust the objective function of the NMF model according to the non-negative matrix Z, and determine the corresponding iterative optimization strategy; Among them, in the non-negative matrix Z, 1 is used to represent missing values, and 0 is used to represent non-missing values; S3, solving the NMF model based on the iterative optimization strategy. In each iteration, other matrices are fixed and one matrix is ​​updated until the iteration termination condition is met, and the decomposed non-negative matrix W, non-negative matrix H and non-negative matrix Z are obtained; Among them, the non-negative matrix W is the basic image matrix, which contains the features or basis extracted from the non-negative matrix V, and each column of the matrix is ​​regarded as a basic feature or basis vector; The non-negative matrix H is a coefficient matrix, which contains coefficients for linearly combining basis vectors in the non-negative matrix W. Each row of the matrix corresponds to a data point in the non-negative matrix V, and each column of the matrix corresponds to a basis vector in the non-negative matrix W. S4, filling missing values ​​in the preprocessed passenger distribution coefficient and vehicle data according to the decomposed non-negative matrix W, non-negative matrix H and non-negative matrix Z, and calculating the filling data corresponding to the missing values; S5. Smooth and verify the calculated filling data to ensure the accuracy and rationality of the filling data.

7. The intelligent subway monitoring system according to claim 6 is characterized in that: A non-negative matrix Z representing the missing value data structure is introduced in S2. The objective function of the NMF model is adjusted according to the non-negative matrix Z, and the corresponding iterative optimization strategy is determined, including: S21, determine the data structure of missing values, and introduce a non-negative matrix Z that represents the data structure of missing values; S22. Adjust the objective function of the NMF model according to the non-negative matrix Z: Among them, V, W, H, and Z are non-negative matrices V, W, H, and Z, respectively. nmv , H nmv are the non-missing values ​​observed in the non-negative matrix V and the non-negative matrix H, respectively, λ is the regularization parameter, represents the Hadamard product, represents the square of the Frobenius norm; S23. Use the alternating minimization strategy as an iterative optimization strategy for solving the NMF model.

8. The intelligent subway monitoring system according to claim 7 is characterized in that: In S3, the NMF model is solved based on an iterative optimization strategy. In each iteration, other matrices are fixed and one matrix is ​​updated until the iteration termination condition is met. The decomposed non-negative matrix W, non-negative matrix H and non-negative matrix Z are obtained, including: S31, solving the NMF model based on the alternating minimization strategy; S32, in each iteration process, the non-negative matrix W, the non-negative matrix H, the non-negative matrix Z, and the padding data are updated in the order of updating; S33, judging whether the iteration termination condition is satisfied, if not, returning to S31, otherwise taking the current non-negative matrix W, non-negative matrix H and non-negative matrix Z as the decomposed non-negative matrix W, non-negative matrix H and non-negative matrix Z; The iteration termination conditions include the objective function value converging to below a certain threshold, or the number of iterations reaching a preset maximum number.

9. The intelligent subway monitoring system according to claim 5, characterized in that: The control terminal uses the control decision module to evaluate the passenger flow and vehicle operation according to the processed received data, and formulates a control scheduling strategy, and sends the control scheduling strategy to the vehicle terminal through the ground base station, including: The control terminal uses the control decision module to evaluate the passenger flow based on the processed passenger distribution coefficient and the heat source area and the number of heat source targets collected by the infrared sensor in the vehicle data in each compartment; The control terminal uses the control decision module to input the operating status data of each compartment collected by other sensors in the processed vehicle data, as well as the corresponding position information when other sensors obtain the operating status data, into the pre-trained vehicle operation status judgment model to evaluate the vehicle operation status; The control decision module conducts a comprehensive analysis of the passenger flow assessment results and the vehicle operation assessment results, and formulates a control scheduling strategy. The control terminal sends the control scheduling strategy to the on-board terminal through the ground base station.

10. The intelligent subway monitoring system according to claim 9, characterized in that: The control terminal uses the control decision module to input the operating status data of each compartment collected by other sensors in the processed vehicle data, as well as the corresponding position information when other sensors obtain the operating status data, into the pre-trained vehicle operation status judgment model, before evaluating the vehicle operation status, including: S1. Divide the historical data set into a training set, a validation set, and a test set according to a preset ratio; S2. Setting the loss function and optimizer of the vehicle operation condition judgment model; S3, inputting the training set into the vehicle operation condition judgment model for model training; S4, the loss value is calculated based on the loss function, and the optimizer updates the model parameters according to the loss value and network gradient information; S5. If the loss value is less than the preset threshold, the model training ends, and the current vehicle operation condition judgment model is the trained vehicle operation condition judgment model. Otherwise, return to S3 and continue to use the training set for model training. S6. Input the validation set into the trained vehicle operation condition judgment model, evaluate the generalization ability of the model by observing the performance on the validation set, and tune the hyperparameters and structure of the model; S7. Input the test set into the optimized vehicle operation condition judgment model to evaluate the model performance.

Images